SkillEvo: 來自多輪互動回饋的自我更新演化梯度

摘要

智能體技能目前要不是由人工編寫，便是在單次大型語言模型（LLM）生成過程中所產生，因此缺乏可供其從自身實際引發之互動失敗中改進的閉環。近期研究確實形成了閉環，但其回饋來自單回合問答評估，其後果是嚴重的失衡：一旦第一輪修補了單次互動所能揭露的缺口，演化梯度便衰減；唯有跨越多回合才會浮現的缺陷依然隱而未現，演化因而停滯。這些系統中的治理同樣由端到端驗證分數驅動；該分數是一道標量閘門，能拒絕劣化的候選者，卻既無法定位也無法修復其結構性成因。我們主張，限制技能持續演化的關鍵約束既不是編輯能力，也不是迭代次數，而是評估回饋能否持續供給可信的演化梯度。我們提出 SkillEvo，其中可信回饋產生梯度，可控治理則約束其方向。第一個組成部分將多回合使用者模擬從評估端點重塑為回饋生成器：追問逐層揭露缺陷，使每一輪修訂既消耗回饋，又產生新的回饋。第二個組成部分以獨立治理層取代標量閘門的被動拒絕，主動修復事實性退化與結構膨脹，防止梯度在退化累積時漂移。在六類雲端服務、9 個生產級技能，以及 98 個技能參考檔案中，SkillEvo 比基於自我反思的演化高出 23.0 分，也比單回合問答驅動的演化高出 15.4 分。

English

Agent Skills are today either hand-authored or produced in a single LLM generation pass, and consequently possess no closed loop through which they might improve from the interaction failures they actually cause. Recent work does close this loop, but derives its feedback from single-turn question-answering evaluation. The consequence is a sharp asymmetry: once the first round has patched the gaps that a single exchange can reveal, the evolution gradient decays, the defects that surface only across multiple turns remain invisible, and evolution stalls. Governance in these systems is likewise driven by an end-to-end verification score, a scalar gate that can reject a degraded candidate but can neither localize nor repair its structural cause. We argue that the binding constraint on sustained skill evolution is neither editing capability nor the number of iterations, but whether the evaluation feedback keeps supplying trustworthy evolution gradients. We introduce SkillEvo, in which trustworthy feedback generates the gradient and controllable governance constrains its direction. The first component recasts multi-turn user simulation from an evaluation endpoint into a feedback generator: follow-up questions expose defects layer by layer, so that every round of revision both consumes feedback and produces new feedback. The second replaces the passive rejection of a scalar gate with an independent governance layer that actively repairs factual degradation and structural bloat, preventing the gradient from drifting as degradation accumulates. Across six categories of cloud services, 9 production Skills, and 98 skill-reference files, SkillEvo surpasses self-reflection-based evolution by 23.0 points and single- turn-QA-driven evolution by 15.4 points.